Reactions of capital markets to financial reporting

1. REACTIONS OF CAPITAL MARKETS TO FINANCIAL REPORTING
LEARNING OBJECTIVES
Upon completing this chapter reader should:
 Understand the role of capital market research in assessing the information content of accounting disclosures;
 Understand the assumptions of market efficiency adopted in capital market research;
 Understand the difference between capital market research that looks at the information content of accounting disclosures, and capital market research that uses share price data as a benchmark for evaluating accounting disclosures;
 Be able to explain why unexpected accounting earnings and abnormal share price returns are expected to be related;
 Be able to outline the major results of capital market research into financial accounting and disclosure
Opening issues
Assume that these are five companies from the same industry with the same balance date, 31 December. All five companies are to make earnings announcements for the financial year (the announcements being made in February), but the earning announcement are spread over two weeks, with no two companies announcing their earnings on the same date.
(a) Would you expect the earning announcements made by each company to impact on their share price, and if so, why?
(b) If it is found that are share price of some entities change more around the date of the earning announcement that others, what might have caused this price-effect differential?
(c) Would you expect the share price of large companies, or smaller companies, to be relatively more impacted by earning announcement?
(d) Ones the first company in the sample of five makes its earnings announcement, would you expect this announcement to impact on the price of share in the other four companies? Why?
Introductions
In some of the previos chapters we have considered various normative perceptions pertaining to how accounting should be undertaken. For example, chapter 4 discussed theories that had been developed in prescribe how accounting should be undertaken in time of rising price (for example, general price level accounting current cost accounting and continuously contemporary accounting). Chapter 5 considered the role of conceptual frameworks in providing prescription (such frameworks can tell us what the objective of accounting is; what qualitative charecteristics accounting information should prosess; how elements of accounting whould be difened and recognized and how assets and liabilities should be measured). Chapter 9 privided an insight into various approaches adopted to disclose information about an organisation’s social and

2. enviromentntal performance (whice has indeed become an area of accounting research that has grown rapidly in recent years)
While the abovementioned chapter provided a great deal of prescription, they tended not to prove any theoretical arguments as to the motivations for managers to make the disclosures. This vold was filled by chapter 7 and 8 which provided different theoretical perspectives about what drove management to make the disclosures. Chapter 7 discussed positive accounting theory and it indicated that where management had a choice in selecting a particular approach to accounting, both efficienty arguments and opportunistic argument could be advanced to explain and predict management’s accounting choice, chapter 8 provided alternative explanations of management behavior. It showed that the choice of a particular accounting method might be made to restore the legitimacy of an organization (from legitimacy theory), or because such disclosure was necessary to retain the support of powerful stakeholders (from stakeholder theory).
While the above material provide a perspectibe of what motivates managers to provided particular accounting information, the material did not consider the futher issues of how individuals, or groups of individuals in aggregate, react to accounting dislosures. This chapter and chapter 11 provide material that address this issue
This chapter and chapter 11 examne the impac of financial accounting and disclosure decision on users of financial report. Specifically, we look at research that forces on the impact of alternative accounting and disclosures choice on the investment decisions of financial statement users such as sharemarket investor, financial analysis, bank leading offecers, and auditors.
Reported profit depends on many financial accounting decisions, managers have much scope in selecting between alternative accounting methods and accounting assumtions. For example, they will choose between expensing or capitalizing particular cost; they will choose between alternative accounting methods such as straight-line or reducing balance depreciation; using ful life of assets to be depreciated; and so on futher, decisions much be made in relationto how much information to disclosure, the medium for disclosure, and, in some circumstances, whether to recognize particular items in the financial statements, or merely disclose them in the footnotes to the statements
Financial reporting decision impact on the information provide to users of financial reports, this in turn may have implications for the decision that user make. There are two assets the impacts of financial reporting decisions: (a) determine the impact of the information on the decisions of individual of information users (behavior research), and (b) determine what impact the relase of information has on share price (capital market research). In this chapter we considere capital market research (which considers reactions at an aggregate or market level). In this chapter 11 we review behavioural research undertaken at the individual level
AN OVERVIEW OF CAPITAL MARKET RESEARCH

3. Capital market research explores the role of accounting and other financial information in equality markets. This ty of research involves examining statistical relations between financial information, and share price or returns. Reactions of investors are evidenced by their capitalmarket transactions. Favourable reactions to information are presumed to be evidenced by price increase in the particular security, wheareas unfovourable reactions to informatins are the particular security whereas unfavourable reactions to information are evidenced by a price decrease. No price change around the time of the release of information implies no reaction to the information (the release does not provide anything that is new).
Conclusions about the market reaction to particulay information releases or events are generally based in evidence from a large number of companies, with data spanning several years. This type of research is often used to examine equity market reactions to announcement of company information, and to assess the relevance of laternative accounting and disclosure choices for investor. If security price change around the time of the release of particular information, and assuming that the information and not some others event caoused the price change, then it is considered that the information was relevant and useful for investment decision making.
In contrast to behavioural research (considered in chapter 11) which analyses individual responses to financial reporting. Capital market research assesses the aggregate effect of financial reporting, particularly the reporting of accounting earning, on investor, by analysis share price decisions is capture in aggregate. But when considering such research, a possible question that come to mind is why have so many research studies been undertaken that focus on the market’s response to accounting earning announcements? Brown (1994. P. 24) provides one answer to this issue. He argues:
For reason are that, accourding to the financial accounting standar board information about earning and its components it the primary purpose of financial reporting arning are oriented toward the intercsts of shareholders who are an important group of financial statement suser; earning is the number most analysed and forecast by security analysis and reliable data on earnings were readily available.
Another important difference between capital market and behavioral research is that capita market research considered only investor, while behavior research is often used toexamine decision making by other types of financial statement users such as bank managers, loan officers or auditor
Capital market research relies on the underlying assumption that equity markets are efficient. Market efficiency is defened in a accordance with the efficient market hypothesis (EMH) an a market that adjust rapidly to fully impound information into share price when the information is released (Fama et al., 1969). Capital arket research in accounting assumes that equity markets are semi-strong from efficient. That is, that all publicly available disclosures, is rapidly and fully impounded into share price in an unbiased manner as it released. Relevant information is not ignored by market.

4. semi-trong-form effecienty is the most relevant for capital market research in accounting, since it relates to use of publicly available information. Other hypotheses about matket efficiency are week form efficiency perspective, and the strong-form-efficiency perspective, this week of information about past price and trading volumes. This strong form of market efficiency assumes that security price, on average, reflect all information known to anyone at that point in the time (including information not publicly available)
Anggi kosong
BAGIAN RENDY HAL 365-368 DEGAAN
Equation 3 shows that total accrual returns can be devided into normal (or expected) returns, given market-wide price movements (αit + ᵦitRmt), and abnormal (or unexpected) returns due to firm- specific share price movements (μit). Normal returns are expected to vary from company to company, depending on their level of systematic risk in relation to the market, while abnormal returns are expected to vary from company to company, depending on whether there is new information about the company that cause investors to revise expectations abour future earnings.
The market model is used to control for share price movements due to market-wide events, allowing the researcher to focus on share price movements due to firm-specific news. For example, part of the 2,6% return earned by CMR Company upon announcing its annual earnings may be do to an overall rise in the market on the announcement day. A researcher analysing the impact of the earnings announcement would control for the impact of this rise in the market by deducting it from CMR Company’s return. Assuming αCMR = 0, ᵦCMR = 1, and Rm = 1%, this calculation would leave an abnormal return (μCMR ) of 1,6%. It is abnormal returns, or firm-specific share price movements, that are analysed by researchers to determine the information effects of company announcements.
Capital market research into the earnings/return relation analyses firm-specific price movements (abnormal returns) at the time of earnings announcements. These abnormal returns are used as an indicator of the information content of the announcements. That is, how much, if any, new information has been released to the capital markets. If there no price reaction, it is assumed that the announcement contained no new information. That is, the information was already known or anticipated by market participants.
Results of Capital Market Research into Financial Reporting

5. Capital market research has been a major focus on financial accounting research over the past thirty years. The research has investigated the information content of earnings as well as many other accounting and disclosure item. Result of this research are useful for both practising accountants and finance professionals such as security analysts. Knowledge of these results is considered to be particularly useful in relation to making financial reporting decisions. More informed choice between accounting and disclosure alternatives can be made if the expected impacts on share prices are anticipated when making financial reporting decisions. A summary comprising some of the more important capital market research result follows.
Hstorical cost income is used by investors
Ball and Brown (1968), in the first major capital market research publication in accounting, investigated the usefulness of accounting earnings under a historical cost model. Prior to their research, there was a widely held view that historical cost accounting methods resulted in “meaningless” information that was not useful for investors and other users of financial statements. Ball and Brown saw the need for empirical evidence about whether accounting earnings, calculated using historical cost accounting principles, provide useful information to investors. They state (p. 159):
If, as the evidence indicates, security prices do in fact adjust rapidly to new information as it becomes available, then changes in security prices will reflect the flow of information to the market. An observed revision of stock prices associated with the release of the income report would thus provide evidence that the information reflected in income numbers is useful.
Using data for 261 US companies, they tested whether firms with unexpected increases in accounting earnings had positive abnormal returns, and firms with unexpected decreases in accounting earnings had negative abnormal returns (on average). Unexpected earnings were calculated (quite simplistically) as the difference between current earnings and previous year earnings. That is, they assumed that this year’s earnings were expected to be the same as last year’s earnings. Monthly share price data were used, with the market model being used to calculate abnormal returns for each company. Cumulative abnormal returns (CARs) were then calculated for each of (a) the full sample of companies, (b) firms with unexpected increases in earnings (favourable announcements), and (c) firms with unexpected decreases in earnings (unfavourable announcements), by summing the average abnormal returns for each of these groups over time.

6. They found evidence to suggest that the information contained in the annual report is used in investment decision making despite the limitations of the historical cost accounting system. This result is evidenced by the CARs during the month of the earnings announcements (month 0). As can be seen from Figure 10.1, firms with unexpected increases in earnings (favourable announcements), represented by the top line in the chart, had positive abnormal returns, while firms with unexpected decreases in earnings (unfavourable announcements), represented by the bottom line in the chart, had negative abnormal returns during the announcement month (on average). Since Ball and Brown’s early study, this research has been replicated many times using more sophisticated data and research methods. The results appear to confirm the usefulness of historical cost income to investors. This is not to say that the historical cost accounting system is the “most useful”, since a present value or current cost accounting system may be more useful, but it gives some credence to the continued use of historical costs.
Prior to an earnings release, investors obtain much of the information they need from other sources
In addition to confirming the usefulness of the historical cost accounting model, Ball and Brown found that most of the information contained in earnings announcements (85-90%) is anticipated by investors. The gradual slope in the lines (wich represent the cumulative abnormal returns) prior to the earnings announcements (which were made at time 0) in Figure 10.1 provide evidence of this. Anticipation of earnings changes by investors indicates that investors obtain much of the information useful for investment decision making from sources other than annual earnings announcements (perhaps from media releases, analysts releases, information about the industry’s production and sales trends, and so on). This is not surprising given that alternative sources of information such as conference calls to analysts, and press releases, are generally more timely than the annual report, which tends to be issued several weeks after balance date and less frequently than many alternative sources of information. Therefore, we can never expect to produce accounting statements that will tell investors everything they may want to know. That is, provision of all relevant information of investors is not a good basis for regulation or practice. When making financial reporting decisions, it is important to remember that, while accounting appears to be an important sources of information for the stock market, it is not the only source of information.
In finishing our discussion of Ball and Brown (1968) it is perhaps worth nothing that this paper is generally accepted as the most cited academic accounting article. It certainly represented quite a change from previous accounting research, which was predominantly normative. Reflecting on the significance of Ball and Brown (1968), Brown (1994, p. 24) states:

7. A number of reasons have been given to explain the papers major impact on the accounting literature since 1968:
- It was cast in the mould of a traditional experiment: hypothesis, data collection, data analysis, conclusion.
- It expressed a view that ran counter to the critics of GAAP (these critics arguing that historical cost accounting information was meaningless and useless)
- It was an early plea for “empirical research”
- It emphasized the use of data to test a belief
- It adopted an information perspective
- It contained the basic elements of a research design that became a model for future research: the semi-strong form of the EMH was a maintained hypothesis, so the focus was on market behavior around an announcement date, earning predictions were modeled to identify the news in, or what was new about, the earning report, GAAp earnings were compared with a primary version of operating cash flows, and abnormal returns were measured by the Market Model and the CAPM.
- it was a particularly robust experiment, in the sense that it has been replicated for firms with different fiscal years, in different countries, and at different times.
- It gave rise to many papers in related areas.
As Chapter 7 indicates, Watts and Zimmerman (1986) credit the development of Positive Accounting Theory, at least in part, to the early experimental approach adopted by Ball and Brown (1968).
CITRA
As chapter 7 indicates, Watts and Zimmerman (1986) credit the development of Positive Accounting Theory, at least in part , to the early experimental approach adopted by Ball and Brown (1968).
The information content of earnings announcements depends on the extent of alternative sources of information
Research indicates that the information content of earnings varies between countries and between companies within a country. For example, Brown (1970) found that when compared to US markets, the Australian market has slower adjustments during the year, with larger adjustments at the earnings announcement date. This result implies that annual reports were a more important source of information for the Australian capital markets than they were for US capital markets because there were fewer

8. alternative sources of information for Australian companies. This difference in the extent of alternative sources of information is partly due to Australian regulations that required only semi-annual rather than quarterly reporting, and is also a function of differences in average firm sizes between the two countries. Smaller firms tend to have fewer alternative sources of information than larger firms, and are less likely to be followed by security analysts. This difference in the extent of alternative sources of information between smaller and larger firms causes differences in the usefulness of earrings announcements, with these being more useful for smaller than larger Australian firms. Therefore, the extent of alternative sources of information should be considered when making financial reporting decisions. (We return to the issue of the size later in this chapter).
The capital market impact of unexpected changes in earnings depends on whether the change is expected to be permanent or temporary
Following Ball and Brown’s finding that the direction of unexpected earnings changes is positively related to the direction of abnormal returns, further research was conducted into the relation between the magnitude of the unexpected change in earnings, or earnings per share, and the magnitude of the abnormal returns. This relationship
is often referred to as the earnings response coefficient. The results show that this is not a one-to-one relationship. Indeed, some research has shown that the average abnormal return associated with a 1% unexpected change in earnings is only 0.1 to 0.15% (Beaver, Lamberd and Morse, 1980). This relationship varies, depending on whether the change in earnings is expected to be permanent or temporary. Permanent increases are expected to result in increased dividends, and therefore future cash flows, and this implies a change in the value of the company. On the other hand, temporary increases are discounted or ignored, since they are not expected to have the same impact on expected future dividends (Easton and Zmijewski, 1989). While some earnings changes such as those due to one-off restructuring charges are obviously temporary, it is more difficult to determine whether other earnings changes are likely to persist.
Earnings persistence depends on the relative magnitudes of cash and accruals components of current earnings
The accruaal system of aounting differs from the cash basis of accounting owing to differences in when cash flows are recognised in the fianncial statements. Under the accrual ssytem, some items are recognised before the cash flows are received or paid (for example, credit sales and purchases), while others are recognised on a periodic basis (for example, the cost of a fixed asset is recognised over its useful life through periodic depreciation charges). Therefore, the accrual process involves adjusting the

9. timing of when the cash inflows and outflows of a firm are recognised to achieve a matching of revenues and expenses. Earnings, the summary performance measure of the accrual system, has fewer timing and mismatching problems than performance measures based on unadjusted cash flows (for example, cash flows from operatiokns). However, application of the accrual system can be a subjective rather than an objective process and dependeing upon the choice made, many different earnings figures can be achieved. For example, if the reducing balance method of depreciation is chosen over the straight-line method, reported profits will initially be lower, owing to a greater depreciation expense. However, reported earnings will be higher in the later years of the asset’s life due to lower depreciation expense. Sloan (1996) undertook a study to see if share prices behave as id investors simply “fixate” on reported earnings without considering how those numbers have actually been determined (that is, what methods of accounting have been employed). According to Sloan (1996, p. 291):
A meaning test of whether stock orices fully reflect available information requires the specification of an alternative ‘naive’ expectation model, ,against which to test the null of market efficiency. The naive model emplooyed in this study is that incestors ‘fixate’ on earnings and fail to distinguish between the accrual and cash flow component of current earnings. This naive earnings expectation model is consistent with the functional fixation hypothesis, which has received empirical support in capital market, behavior and experimental research.
Sloan provides ecidance that dirms with large accruals relative to their actual cash flows are ulikely to have persistenly high earnings, since the accruals reverse over time, reducing future earnings. However, share prices are found to act as if investors simply fixate on reported earnings, threby failing to take account of the realive magnitudes of the cash and accrual components of current earnings (fixation implying a degree of market inefficiency). In concluding his paper, Sloan (p.314) states:
This papaer investigates whether stock price reflect information about future earnings contained in the accrual and cash flow components of current earnings. The persistence of earnings performance is shown to depend on the realive magnitudes of the cash and accrual components of earnings. However, stok price act as if investors fail to identify correctly the different proporties of these two components of earnings.
Hence, while earnings can be managed up through various discretionary accruals, this cannot be done indefinitely, and earnings will eventually be lower as the accruals subsequently reverse. Likewise, while it may be possible to increase share price through reporting higher earnings, this effect will be reversed when lower earnings are reported in the future. Although the lower earnings are due to the reversal of accraulas, evidence indicates that the market fixates on the lower reported earnings, and share prices therefore fall as the accruals subsequently reverse. Notiions of market efficiency, such as the functional fixation perspective, have been hotly contested in the accounting literature.

10. The earnings announcements of other firms in the same industry have information content
When a companu announces its annual earnings, this generally results in abnormal returns for not only the company conccerned, but also fot other companies in the same industry (Foster, 1981). That is, there is information contetnt for similar firms as well as for the announcing firm. This phenomenon, known as ‘information transfer’, reduces the surprise (unknown) element in earnings announccements of other firms in the industry who choose to announce their earnings later. The direction of the capital market reaction is related to whether the news contained in the announcement redlects a change in conditions for the entire industy, or changes in realive market share within the industry. Because information is gained from the announcements of similar firms, information releases about sales and earnings changes result in pricce reactions for other dirms in an industry, as well as the firm making the announcement (Freeman and Tse, 1992). Therefore, in forming expectations about how information releases, such as earnings announcements, might affect share prices, it is important to consider the timing of information releases relative to those of similar firms, since information about a company can be gained from the information releases of similar companies.
Firth (1976) investigated the information transfer issue. He sought to investigate the impact of a company’s results being publicly announced on the share price behaviour of competing firms’ (p. 298). His results indicated that when ‘good news’ was released about accounting earnings, the sahre prices of the non-announcing firms in the same industry reacted quickly (the same day) by showing a statistically significant increase. Interestingly there seemed to be no abnormal return in the days before announcement (limited anticipation) or the days after the announcement. That is, most of the adjustment appeared to occur fairly immediately. Simimlar results were found with respect to ‘bed news’ earnings announccements. The, share prices of nondisclosing firms in the industry tended to fall on the day of the ‘bad news’ earnings announcement.
If the announcement of earnings by one company impacts in the share prices of other companies in the same industry, there would be an expectation that if a number of companies are about to release earnings information , then all other things being equal, the largest share price reactions might be generated by the entity that makes the first release. By the time the last entity releases its earnings announcements for a particular year end, it could be expected that a great deal of this information would already have been impounded in share prices and hence the last announcement woul have relatively little impact on share prices. This expectation was confirmed by Clinch and Sinclair (1987). In summarising their findings they state (p.105):

11. The directional association between daily proce changes for announcing and non-announcing firms and the magnitude of the price change diminishes for subsequent announcing firms in the same industry over the reporting period.
We can relate the view that annoucnements by one company can impact on share price of another company to a recent newspapaer article relating to banks. Exhibit 10.1 shows that when one banl released its profit figures, this seemed to cause share proce reactions in the shares of other banks. This is consistent with the result reported above.
Earnings forecasts have information content
Announcements of earnings forecasts by both management and security analysts are associated with share returns. That is, not only do announcements of actual earnings appear to cause share prices to change, but annoincements of expected earnings also appear to cause price changes. Similar to earnings announcments, earnings forecasts are associated with market return in terms of both direction and magnitude (Penman, 1980; Imhoff and Lobo, 1984). These results are not surprising since earnings forecasts are expected to contain new information that can be used in the prediction of future earnings. Forecasts of expected future earnings appear to be an effectivve way of communicating information to the share market. Also, bad news forecasts about lower than anticipated future earnings may be usefu; for avoding potential shareholder lawsuits (Skinner, 1997).
Earnings forecasts have also been explored in terms of the information transfer phenomenoon discuseed above. Baginski (1987) generated results that showed that the share prices of firms within tha same industry that did not provide an earnings forecast were positively correlated with the change in earnings expectation indicated by earnings forecasts released by managers of other firms within the same industry.
Exhibit 10.2 provides an example of where it appears that a profit forecast led to an increase in share price, consisent with some of the research results reported abouve. However, what must be remembered in research that looks at market reaction around particular events is that is possible that share prices might actually be reacting to other unknown (by the researcher) contemporaneous events.
ARIESTA
Prices might actually be reacting to other unknown (by the researcher) contemporaneous events.
There are benefits associated with the voluntary disclosure of information
The disclosure of additional information, over and above that required by accounting regulations, has benefits in the capital markets. Voluntary accounting disclosure include those contained within the annual report, as well as those made via other media such as press released and conference calls to

12. security analysts. For example, Lang and Lundholm (1996) show that firms with more informative disclosure policies have a larger analyst following and more accurate analyst earnings forecasts. They suggest that potential benefits to disclosure include increased investor following and reduced information asymmetry. Further, Botosan (1997) shows that increased voluntary disclosure within the annual report is associated with reduced costs of equity capital, particularly for foirms with low analyst following. For these low analyst firms, disclosure of forecast information and key non-financial statistics is particularly important, while the firms with a high analyst following, disclosure of historical summary information is beneficial. Therefore, these potential benefits of increased disclosure should perhaps not be ignored when deciding the extent of voluntary disclosure.
Recognition is perceived differently to mere footnote disclosure.
Recognising an item by recording it in the financial statements and including its numerical amount in th efinancial statement’s totals is perceived differently to merely disclosing the amount in footnotes to the statements. For example, Abody (1996) finds that where firms in the US oil and gas industry recognise a write down in their financial statement, the negative pricing effect is significant. In other word, the market perceives recognised write downs to be more indicative of a value decrement than disclosure write downs. Further, Cotter and Zimmer (1999) show that mere disclosure of current values of land and buildings indicates that the amount is less certain when compared to current values that are recognised in the statement of financial position (balance sheet) via an asset revaluation. The results indicate that investors place greater reliance on recognised amounts than on disclosed amounts. Therefor, while it is not not necessary to recognise information in the balance sheet for it to be useful, since merely disclosing the information in the footnotes conveys the information to investors. Discloded information is perceived to be less reliable than recognised information.
Size
There is evidence that the relationship between earnings announcements and share price movements is inversely related to the size of an entity. That is, earnings announcements have been found generally to have a greater impact on the share prices of smaller firms relative to larger firms. This is explained on the basic that with larger firms there is generally more information available in the marketplace and therefore greater likelihood that projections about earnings have already been impounded in the share price. The earnings announcement for larger firms would have a relatively limited unexpected component. For example, research har shown that the relationship between the information content in earnings announcements and changes in share prices tends to be more significant for smaller firms.that is, in general, larger firms’ earnings announcemnet hane relatively less information content (for example, Freeman, 1987; Collin, Kothari and Rayburn, 1987). This is consistent with the EMH and it explained by the fact that larger firms tent to have more attention from such paties as security analysts. Hence, on average, earnings announcements for larger firms tent to be more anticipated and hence already impounded in the share price prior to the earnings announcement.
Grant (1980) alsp explored this issue. He investigated the information context of earnings announcements of securities trede on the New York Stock Exchange, as well as securities traded on what is known as the Over The Counter Market (OTC). Firms traded on the OTC are typically smaller than firms traded on the New York Stock Exchange. Consistent with the perceived size effect, the price of securities traded on the OTC were more resposive to earnings announcements thanwere the prices of securities traded on the New York Stock Exchange.

13. Do current share anticipate future accounting earnings announcement?
The previous discussion in this chapter has provided evidence that accounting earnings announcements, because of their potential information context, can have some impact on share prices. That is, prices change in relation to information as it becomes available. However, this impact seems to be more significant for smaller firms. As we saw, this is explained on the basis that there tends to be more information available for larger firms (for example, through analysts paying particular attention to the larger firms). Hece, as firm size increases, the general perspective taken is that shre prices incorporate information from a wider number of sources (incliding, perhaps, numerous forecasts of the larger entity’s earnings) and therefore there is relatively less unexpected information when earnings are ultimately announced. Therefore, for larger firms, we might actually be able to argue that share prices anticipate future earnings announcements with some degree of accuracy. As Brown (1994, p. 105) states, if we take the perspective that share prices anticipate earnings announcements, than we are effectively ‘looking back the other way’ from traditional perspectives that assume thats earnings announcement actually drives share price changes.
A more recent focus in capital market research investigates how well accounting information, such as annual earnings, captures information that is relevant to investors. This is a different focus from research considered previously in this chapter. Rather than determining whether earnings announcment provide information to investors, this alternative form of research seeks to determine whether earning announcement reflect information that has already been used by investors in decision making. That is, this research views market prices, and hence returns, as leading accounting earnings, while ‘information content’ research (the previous focus of this chapter) views earnings as leading (or driving) market returns. Both perspectives have merit, since the earnings announcement is likely to contain some information about a firm’s activities that was not previously known to investors, as well as information that investors have already determined (or anticipated) from alternative sources.
In concidering why share prices convey information about future accounting earnings, brown (1994, p. 106) argues :
In a world of rational expectations, events that affect future distributions to shareholders will be reflected in today’s share price, whereas accounting standards often require that the recognition of those events be deferred until some future accounting period.
ENDANG
Shares prices and returns (changes in prices plus dividends) are considered by some researchers to provide useful benchmarks for determining whether accounting information is relevant for investor decision making. Share prices are deemed to represent a benchmark measure of firm value (per share), while share returns represent a benchmark measure of firm performance (per share). These benchmarks are in turn used to compare the usefulness of alternative accounting and disclosure methods. For example, this methodology is used to answer questions such as: are cash flows from operations a better measure of a firm’s performance than earnings calculated using the accruals system? Each of these accounting measures of performance is compared to the market benchmark measure of performance

14. (returns) to determine which accounting measure best reflects the market’s assessment of company performance. This type of capital market research also assumes that the market is efficient, and acknowledges that financial statements are not to the only source of information available to the markets, such that security prices reflect information that is generally available from a multitude of sources. In particular, it assumes that investor and financial analysts actively seek out relevant information when making investment decisions or recommendations, rather than awaiting the release of the annual report. Further, this area of the research allows researchers to consider questions about statement of financial position measures. For example, are disclosures about current values of assets value-relevant? That is, are they associated with, or linked to, the current market value of the company? Again, there is an assumption that market value s reflect all publicly available information, including but no limited to information contained in financial statements.
This area of the research is based on a theoretical framework that is derived from the premise that market values and book values are both measures of a firm’s value (stock of wealth) even though book value measures wealth with some error. That is, at any point in time, the market value of a company’s equity (MV) is equal to the book value of shareholder’s equity (BVit) plus some error (ɛit) :
MVit = BVit + ɛit
This error is due to the conservative nature of the accounting system. Book value is generally expected to be lower than market value for a number of reasons. First, not all assets and liabilities are recognized in the financial statements. For example, human resources, customer satisfaction levels, and internally generated goodwill are not included in the statement of financial position, nor are their values amortized to the statement of financial performance (income statement). Second, some assets are recognized at less than their full value. For example, fixed assets, that have not been revalued, and inventory are generally recorded at less than their expected sale prices.
If markets are assumed to be efficient, market value provides a benchmark measure against which alternative measures of book value can be assessed. As we see shortly, there is a deal of research that evaluates the output of the accounting system on the basic of how the accounting information relates or compares to current market prices of the firm’s securities.
If market value and book value of a company are considered as ‘stocks’ of wealth, then changes in each of these measures of wealth between two points in time can be considered as ‘flows’ of wealth. Just as market of book measures of stocks of wealth equate with error, market and book measures of flows of wealth (changes in value) can be equated, albeit with some degree of error:
ΔMWit = ΔBVit + ɛ it …………………… (Equation 5)

15. Change in market value (ΔMWit) is simply the difference in the market capitalization of a company between two points in time (t – 1 to t). on a ‘per share’ basis, it can be expressed as the change in the price of one share:
ΔMWit / no. of shares = Pit - Pit – 1 ……. (Equation 6)
Change in book value (ΔBVit) is the difference between opening and closing total shareholder’s equity. However, if we assume that there have been no additional capital contributions during the period, ΔBVit can also be measured by considering the change in retained earnings for the period. On a per share basis, this is measured as earnings per share (Eit) less dividend paid per share (Dit):
ΔBVit / no. of shares = Eit - Dit ………… (Equation 7)
This formula is based on the concept of ‘clean surplus’ earnings, which assumes that all increases book value pass through the statement of financial performance. Clean surplus earnings does not always hold in practice, since items such as asset revaluation increments are credited directly to owner’s equity. However, the assumption of clean surplus is useful for simplifying our analysis. Substituting Equations (6) and (7) into Equation (5) gives:
Pit - Pit – 1 = Eit - Dit + ɛ it ………………. (Equation 8)
That is, there is a theoretical relationship between change in price and change in retained earnings for the period. With a small amount of manipulation, this equation can be expressed to relate returns to earnings (the return/earnings relation). First, adding dividends to both sides of the equation, and dividing through by beginning of period prices, gives:
= Eit / Pit-1 + ɛ’’it ……………….. (Equation 9)
Since the left-hand side of the equation, ((
/
, is equal to returns (equation 2), we are left with an equation relating returns to earnings:
Rit = Eit / Pit-1 + ɛ’’it ……………………………….. (Equation 10)
Equation 10 shows that we should expect returns (Rit) and earnings per share divided by beginning pf period price (Eit / Pit-1) to be related.
In short, this perspective says that if market value is related to book value, returns should be related to accounting earnings per share, divided by price at the beginning of the accounting period. This analysis provides an underlying reason why we should expect returns to be related to earnings over time. However, it is interesting to note that it is total earnings per share rather than unexpected earnings per share that this theoretical framework proposes should be

16. associated with returns. This is in contrast to research that assesses the ‘information content’ of earnings announcements by analyzing the association between unexpected earnings per share and abnormal returns at studies that evaluate reported earnings on the basis of how closely the movements in reported earnings (or EPS) relate to changes in share prices.
Beaver, Lambert and Morse (1980) was an early paper that sought to investigate how efficiently data about share prices enable a researcher to estimate future accounting earnings. Accepting that share price is the capitalized value of future earnings, they regressed the annual percentage change in share price on the percentage change in annual EPS. Consistent wit Equation 10, they found that share prices and related returns were related to accounting earnings, but they also found that share prices in year t were positively associated with accounting earnings in year t + 1 (share prices led accounting earnings). It was accepted that share price movements provided an indication of future movements in accounting earnings. Because of various information sources, prices appeared to anticipate future accounting earnings. These findings were also supported in a later study by Beaver, Lambert and Ryan (1987) who regressed changes in security prices on the percentage changes in earnings.
In previous discussion we showed that research indicates that share prices of larger firms do not adjust as much to earnings announcements as do the share prices of smaller firms. This was explained on the basis that there is more information available and analyzed in relation to larger firms, and hence information about earnings is already impounded in the share price. That is, ‘for larger firms, there is a broader and richer information set, and there are more market traders and more analysts seeking information’ (Brown, 1994, p. 110).
Now if we adopt the position taken in this section of the chapter that share prices can actually anticipate earnings announcements (‘looking back the other way’), as indicated in Beaver, Lambert and Morse (1980) and Beaver, Lambert and Ryan (1987), then perhaps share prices anticipate accounting earnings more efficiently in the case of larger firms. Collins, Kothari and Rayburn (1987) found evidence to support this view-size does matter, with the share prices being a better indicator of future earnings in larger companies.
As noted above, there have been a number of studies using stock market valuations as a basis for evaluating accounting information. In Equation 10 we indicated that, theoretically, market returns should be related to earnings. Dechow (1994) investigates how well accounting earnings reflect market returns. She also considers whether another measure of performance, based on cash flows, relates better to returns than earnings based on the accrual accounting system. According to Dechow (p. 12):
This paper assumes that stock markets are efficient in the sense that stock prices unbiasedly reflect all publicly available information concerning firm’s expected future cash flows. Therefore, stock price performance is used as a benchmark to assess whether earnings or realized cash flows better summarize this information.

17. According to Dechow, earnings are predicted to be a more useful measure of firm performance than cash flows because they are predicted to have fewer timing and matching problems. In her conclusions, she states (p. 35):
This paper hypothesizes that one of accounting accruals is to provide a measure of short- term performance that more closely reflects expected cash flows than do realized cash flows. The result are consistent with this prediction. First, over short measurement intervals earnings are more strongly associated with stock returns. This is in contrast to research that assesses the ‘information content’ of earnings announcements by analyzing the association between unexpected earnings per share and abnormal returns at the time of the announcement. We will see that there has been a number of studies that evaluate reported earnings on the basis of how closely the movements in reported earnings (or EPS) relate to changes in share prices.
Beaver, Lambert and Morse (1980) was an early paper that sought to investigate how efficiently data about share prices enable a researcher to estimate future accounting earnings. Accepting that share prices is the capitalized value of future earnings, they regressed the annual percentage change in share prices on the percentage change in annual EPS. Consistent with Equation 10, they found that share prices and related returns were related to accounting earnings, but they also found that share prices in year t were positively associated with accounting earnings in year t + 1 (share prices led accounting earnings). It was accepted that share price movements provided an indication of future movements in accounting earnings. Because of various information sources, prices appeared to anticipate future accounting earnings. These findings were also supported in a later study by Beaver, Lambert and Ryan (1987) who regressed changes in security prices on the percentage change in earnings.
In previous discussion we showed that research indicates that share prices of larger firms do not adjust as much to earnings announcements as do the share prices of smaller firms. This was explained on the basis that there is more information available and analyzed in relation to larger firms, and hence information about earnings is already impounded in the share prices. That is, ‘for larger firms, there is a broader and richer information set, and there are more market traders and more analysts seeking information’ (Brown, 1994. P. 110).
Now, if we adopt the position taken in this section of the chapter that share prices can actually anticipate earnings announcements (‘looking back the other way’), as indicated in Beaver, Lambert and Morse (1980) and Beaver, Lambert and Ryan (1987), then perhaps share prices anticipate accounting earnings more efficiently in the case of larger firms. Collins, Kothari and Rayburn (1987) found evidence to support this view-size does matter, with the share prices being a better indicator of future earnings in larger companies.
As noted above, there have been a number of studies using stock market valuations as a basis for evaluating accounting information. In Equation 10 we indicated that, theoretically, market returns should be related to earnings. Dechow (1994) investigates how well accounting

18. earnings reflect market returns. She also considers whether another measure of performance, based on cash flows, relates better to returns than earnings based on the accrual accounting system. According to Dechow (p.12):
This paper assumes that stock markets are efficient in the sense that stock prices unbiasedly reflect all publicly available information concerning firm’s expected future cash flows. Therefore, stock price performance is used as a benchmark to assess whether earnings or realized cash flows better summarize this information.
According to Dechow, earnings are predicted to be a more useful measure of firm performance than cash flows because they are predicted to have fewer timing and matching problems. In her conclusions, she states (p.35):
This paper hypothesizes that one role of accounting accruals is to provide a measure of short-term performance that more closely reflects expected cash flows than do realized cash flows. The result are consistent with this prediction. First, over short measurement intervals earnings are more strongly associated with stock returns than realized cash flows. In addition, the ability of realized cash flows to measure firm performance improve relative to earnings as the measurement interval is lengthened. Second, earnings have a higher association with stock returns than do realized cash flows in firms experiencing large changes in their working capital requirements and their investments and financing activities. Under these conditions, realized cash flows have more severe timing and matching problems and are less able to reflect firm performance.
Using the market value of a firm’s securities as a benchmark a number of studies have also attempted to determine which asset valuation approaches provide accounting figures that best reflect the valuation the market places on the firm.
ANJENG
Using the market value of a firm’s securities as a benchmark, a number of studies have also attempted to determine which asset valuation approaches provide accounting figures that best reflect the valuation the market places on the firm. The perspective taken is that book values that relate more closely to market values (determined through a review of share prices) provide more relevant information than other accounting valuation approaches. Barth, Beaver and Landsman (1996) undertook a study that investigated whether fair value estimates of a bank’s financial instruments (as required in the US by SFAS No. 107) seem to provide a better explanation of bank share prices relative to values determined on the basis of historical cost accounting. Their findings indicate that SFAS No. 107 disclosures provide significant explanatory power for bank

19. share prices beyond that provided by book values (p. 535), thereby providing evidence that such values are the values that are relevant to investors.
Easton, Eddey, and Harris (1993) investigate whether revaluations of assets result in an alignment between information reflected in annual reports and information implicit in share prices and returns. Again, share price data are used as the benchmark against which accounting data is assessed. According to Easton et al. (p.16)
Prices are used to assess the extent to which the financial statements, including asset revaluations, reflect the state of the firm at a point in time, while returns are used to assess the summary of change in financial state that is provided in the financial statements.
They found that the revaluation of assets generally resulted in better alignment of market and book values. In concluding their paper, they state (p. 36)
Our analyses support the conclusion that book values including asset revaluation reserves are more aligned with the market value of the firm than book values excluding asset revaluation. That is, asset revaluation reserves as reported under Australian GAAP help to provide a better summary of the current state of the firm. Thus, allowing or requiring firms to revalue assets upward should be carefully considered by organizations such as the UK Accounting Standard Board, the Japanese Ministry of Finance, and the International Accounting Standard Committee as they debate the merits of various proposed changes in asset revaluation practice.
Consistent with a view that market prices already seem to reflect the current values of an entiry’s assets, as indicated in the research discussed above, it is interesting to note that asset revaluations do not appear to provide information to investors over and above historical cost accounting information. That is, some studies have indicated that the provision of current cost data in financial statements does not have information content (Brown and Finn, 1980). These results, suggest that investors are able to estimate current value information price to it being disclosed in the financial statements. Therefore, while the provision of current value information does not provide new information to investors, it appears to reflect the information used by investors in making their investment decisions.
Chapter Summary
In concluding our discussion on capital market research we can see that this research has investigated a number of issues. Central to the research are assumptions about the efficiency of

20. the capital market. In the “information content” studies considered in the earlier part of this chapter we saw that researchers investigated share market reactions to the releases of information, often specifically the release of accounting information. The view taken was that the accounting disclosures often revealed new information and in a market that was deemed to be informationally efficient with regard to acting upon new information, share prices react to this information
In the latter part of this chapter we considered studies that investigated whether accounting disclosures reflected, or perhaps confirmed, information already impounded in share prices. The perspective taken was that in an informationally efficient capital market, market prices of shares will reflect information from a multitude of sources. If the accounting information did not reflect the information already impounded in share prices, then some researchers would argue that accounting data that do not relate to share prices, and changes therein, are somewhat deficient. The idea is that market prices reflect information from many sources and if the accumulated data provide a particular signal, the accounting disclosures should also provide similar signals. As we noted, this approach implies that the market has it “right” when determining share prices, and hence returns. In practice, the market cannot be expected always to get it “right”.
Because much knowledge about the performance of an entity will be gained from sources other than accounting, it is perhaps reasonable to expect that accounting information should relate (not perfectly) to expectations held by capital market participants, as reflected in share prices. However, because there will arguably always be some unexpected information released when accounting results are made public, we might expect that not all accounting disclosures will be of a confirmatory nature. Some information in the accounting releases will be new, and in a market that is assumed to be information efficient, some price revisions are to be expected.
Many share price studies have investigated the market’s reaction to particular disclosures. Often, if no reaction is found, it is deemed that the information is not useful and therefore entities should not go to the trouble and expense involved in making such disclosures. This form of argument has been used to criticise accounting regulators for mandating particular disclosure requirements. What must be recognised, however, is that capital market research investigates the aggregate reaction of one group of stakeholders, the investors. While the share market is an important user of accounting information, provision of information to the share market is not the only function of the accounting system. Accounting information is also used for monitoring

21. purposes (Jensen and Meckling, 1976; Watts and Zimmerman, 1990), and hence, financial statements play an important role in relation to the contracting process. Financial statements provide a relatively low cost way of measuring managers performance and monitoring compliance with contract terms, thereby helping to reduce agency costs. Further, financial statement information can be used to satisfy people’s rights to know, that is, to fulfil the duty of accountability. Therefore, while investors are important users of financial statements, it would be foolish to focus solely on the investor-information role of financial reporting, to the exclusion of considerations about the monitoring/ accountability role.